Position-sensitive diffractive imaging in STEM by an automated chaining diffraction algorithm.

The diffractive imaging process used for retrieval of an aberration-free exit-wave function of a complex-valued object is optimized with a newly developed automated chaining diffraction (ACD) algorithm. Our algorithm enables automatic recovery of the amplitude and phase of the complex-valued objects with diffraction-limited resolution, starting from selected-area electron diffraction (SAED) patterns recorded from partially overlapping regions in STEM/CTEM. Based on a 'differential map' (DM) approach, the ACD algorithm meets very general requirements and, similar to 'hybrid input-output' (HIO) algorithm, can be applied to non-periodic, real or complex structures. In contrast to many other algorithms, it is not limited by the object's finite size or tight object support. Wide-field-of-view reconstructions for the complex-object-wave amplitude and phase made with ACD algorithm from SAED patterns down to sub-Angstrom resolution show the potential of diffractive imaging for quantitative analysis of functional materials at different length scales in terms of absorption and scattering mechanisms. The method can be applied also for imaging magnetic properties of samples by the electron or neutron microscopy and/or imaging of non-periodic objects with X-ray microscopy.